scalata.application.usecases.enemyusecases

Executes the enemy attack phase.

==== Algorithm ====

• Fetch the current Player from the session.

• For every alive enemy in the room:

• Compute the four-tile neighbourhood around the enemy (neighboursFiltered).

• If the player’s position lies in that neighbourhood, apply enemy.attack(player).

• Return the player after all damage has been accumulated (purely functional fold).

The method is pure: it produces a new Player instance without mutating the session or the room.

Calculates one AI movement step for every living enemy inside the supplied room.

==== Key steps ====

• Generate Prolog facts – encode the grid size, obstacles (items + other enemies), enemies’ positions and the player location/visibility.
• Run Prolog rules (breadth-first search) – compute the cost-to-player for every reachable tile and derive best_moves_all/1.
• Conflict resolution – group candidate moves per enemy, keep only the cheapest ones, then greedily pick non-overlapping destinations.
• Apply moves – return a new scalata.domain.world.Room with updated enemy positions.

The function is pure: it builds a Prolog engine, computes the moves and returns a modified room; the original room and session remain unchanged.